Featured · Dexterous Robotics

RUKA-v2: Tendon Driven Open-Source Dexterous Hand with Wrist and Abduction for Robot Learning

Xinqi Liu, Ruoxi Hu, Alejandro Ojeda Olarte, Zhuoran Chen, Kenny Ma, Charles Cheng Ji, Lerrel Pinto, Raunaq Bhirangi, Irmak Guzey

Accepted to ICRA Workshop · Spotlight

RUKA-v2 is a fully open-source, tendon-driven humanoid hand for robot learning. It extends the original RUKA hand with a decoupled 2-DOF parallel wrist and finger abduction/adduction, enabling more human-like dexterous manipulation behaviors such as confined-space manipulation, in-hand rotation, and calligraphy.

The system supports bimanual and single-arm teleoperation across dexterous tasks, as well as autonomous policy learning for robot manipulation.

More Coming Soon

Additional papers and projects

I’ll add more projects and publication entries here as I organize my research work, code, videos, and project pages.